The startup that jolted the AI industry with a cheap, open-weight model is moving fast to capitalize on its fame. DeepSeek is in talks to raise about $1.5 billion at roughly a $71 billion valuation, and is preparing for an eventual public listing, according to Bloomberg and the Financial Times on July 14, 2026.
The numbers
Underlying reports peg the new round at at least ~$1.4 billion, potentially several times higher, against a pre-money valuation of about 480 billion yuan (~$71 billion). The investors in this prospective raise are described only as unnamed new backers, with talks still ongoing.
A month after the first round
What makes the timing striking is how recent DeepSeek's last raise was. The company closed its first-ever outside round — $7 billion at roughly a $50 billion valuation — in early June, backed by Tencent, CATL, JD.com, NetEase, IDG Capital and China's National AI Industry Investment Fund. A jump to ~$71 billion barely a month later signals intense demand for exposure to China's most prominent open-model lab.
Toward a listing
DeepSeek is reportedly planning a mainland-China IPO, aiming to finish its financials by the end of December and to file by year-end 2026 or early 2027, with a debut expected in 2027 — though some accounts suggest a filing could come as early as the end of this year. Founded in 2023 and based in Hangzhou, the company has largely eschewed outside capital until now.
The strategic backdrop
DeepSeek's rise has become a symbol of China's open-weight push, with Chinese models increasingly dominant on download charts. A large raise plus a domestic listing would give it the capital to keep training frontier models — and hand Beijing a homegrown AI champion with public-market visibility, at a moment when the government is weighing how freely Chinese open models should be released abroad.